Case 07878847 as the inspiration for this feature request
Background:
Currently, Veeam Backup & Replication for Linux Hardened Repositories uses the default /tmp directory for storing temporary files generated during backup operations. In production environments, especially with large or frequent backup jobs, this can lead to the /tmp filesystem becoming full, resulting in failed or interrupted backup jobs.
Current Workarounds:
Manual cleanup of old files from /tmp.
Increasing the size of the /tmp partition.
Implementing custom cleanup scripts to delete obsolete files.
While these steps help mitigate the issue, they require ongoing manual intervention and do not address the root problem.
Requested Enhancement:
An official feature in Veeam Backup & Replication that allows administrators to configure the location of the temporary directory used by Veeam components on Linux Hardened Repositories. This should include:
A supported and documented setting for specifying an alternative directory (other than /tmp) for temporary files.
Granular configuration, ideally per repository or job, rather than a global setting.
Documentation and guidance on permissions and best practices for the chosen directory.
Benefits:
Reduces risk of backup failures due to /tmp filling up.
Improves flexibility in environments with strict partitioning or storage policies.
Enhances system stability and reduces manual maintenance overhead.
Veeam Infrastructure Appliance has its partitions sized appropriately to avoid any such issues, so this should not be a problem going forward. Also, you should never need to clean up tmp manually, in case you see files get stuck there for extended time with V13 then we would be really interested to investigate what is causing this. In fact, any such issues would be quite critical for us to address ASAP in light of the upcoming "VBR as a service" offering.